Skip to main content
NCBI home page
Clinical and Experimental Immunology logoLink to Clinical and Experimental Immunology
. 2015 Apr 14;180(2):271–279. doi: 10.1111/cei.12571

Value of allohaemagglutinins in the diagnosis of a polysaccharide antibody deficiency

H Schaballie *,†, F Vermeulen *, B Verbinnen †, G Frans †,‡, E Vermeulen †,‡, M Proesmans *, K De Vreese §, MP Emonds †,‡,§, K De Boeck *, L Moens †, C Picard ¶,**, X Bossuyt †,‡, I Meyts *,†,✉
PMCID: PMC4408162  PMID: 25516411

Abstract

Polysaccharide antibody deficiency is characterized by a poor or absent antibody response after vaccination with an unconjugated pneumococcal polysaccharide vaccine. Allohaemagglutinins (AHA) are antibodies to A or B polysaccharide antigens on the red blood cells, and are often used as an additional or alternative measure to assess the polysaccharide antibody response. However, few studies have been conducted to establish the clinical significance of AHA. To investigate the value of AHA to diagnose a polysaccharide antibody deficiency, pneumococcal polysaccharide antibody titres and AHA were studied retrospectively in 180 subjects in whom both tests had been performed. Receiver operating characteristic curves for AHA versus the pneumococcal vaccine response as a marker for the anti-polysaccharide immune response revealed an area under the curve between 0·5 and 0·573. Sensitivity and specificity of AHA to detect a polysaccharide antibody deficiency, as diagnosed by vaccination response, were low (calculated for cut-off 1/4–1/32). In subjects with only low pneumococcal antibody response, the prevalence of bronchiectasis was significantly higher than in subjects with only low AHA (45·5 and 1·3%, respectively) or normal pneumococcal antibody response and AHA (2·4%). A logistic regression model showed that low pneumococcal antibody response but not AHA was associated with bronchiectasis (odds ratio 46·2). The results of this study do not support the routine use of AHA to assess the polysaccharide antibody response in patients with suspected immunodeficiency, but more studies are warranted to clarify the subject further.

Keywords: allohaemagglutinins, bronchiectasis, isohaemagglutinins, polysaccharide antibody deficiency, specific antibody deficiency

Introduction

Specific polysaccharide antibody deficiency (SPAD) is a primary antibody deficiency defined as a poor antibody response to unconjugated pneumococcal polysaccharides with intact responses to protein and conjugate vaccines and normal total immunoglobulin (Ig) concentrations 1. Clinical manifestations of SPAD include recurrent bacterial upper and lower respiratory tract infections, such as sinusitis, otitis media, bronchitis and pneumonia 2,3. SPAD is found in 5–23% of patients who undergo immunological evaluation for recurrent infection 2–9. Antibody deficiency to pneumococcal capsular polysaccharides can also be found as a trait of other primary immunodeficiencies (PID), including common variable immunodeficiency, Wiskott–Aldrich syndrome, ataxia telangiectasia, 22q11.2 deletion, nuclear factor kappa B (NF-κB) essential modifier (NEMO)-deficiency 10 and autosomal dominant hyperimmunoglobulin (Ig)E syndrome 11. Moreover, it is associated with IgG subclass deficiency, selective IgM deficiency, selective IgA deficiency and asplenia. Secondary immunodeficiencies associated with polysaccharide antibody deficiency include splenectomy, immunosuppression and acquired immunodeficiency syndrome 11. In children aged less than 2 years a deficient polysaccharide response is physiological. In young patients (aged more than 2 years) with polysaccharide antibody deficiency, this condition may be transient and represent a delayed maturation of the immune response to polysaccharides 7,12.

A deficient antibody response to polysaccharides is detected by measurement of anti-capsular-polysaccharide IgG antibody concentration before and 2–8 weeks after immunization with unconjugated pneumococcal vaccine (UPV, e.g. Pneumo23®) 12–16. The anti-pneumococcal polysaccharide antibody (Pn antibody) titres are generally assessed by enzyme-linked immunosorbent assay (ELISA). Recently, questions have been raised as to whether or not assessment of pneumococcal antibody response is still a good measure to evaluate the polysaccharide response 12. Previous vaccination with conjugated pneumococcal vaccine may influence the response to UPV, and the increasing number of serotypes included in the conjugated vaccine leaves few serotypes outside the vaccine to be tested 17. This is only one of the problems with interpretation of the current laboratory measurements of Pn antibody response 18. Furthermore, reports on hyporesponsiveness to subsequent vaccinations with polysaccharide vaccines raise concerns on the potential harmful effect of immunizing patients, who may already be at increased risk of infection, with a pure polysaccharide vaccine for diagnostic reasons 19.

Evaluation of the serum titre of allohaemagglutinins (AHA), known formerly as isohaemagglutinins, has been proposed as an alternative to determine polysaccharide responses 12,16,20–22. AHA are antibodies reactive with the A or B polysaccharide antigens on erythrocytes. They were originally called ‘natural antibodies’, or isohaemagglutinins, because they were found without a known history of sensitization. It is now generally assumed that these antibodies are generated in response to polysaccharides on gut bacteria and cross-react with AB blood group antigens 23,24. Because AHA are formed only after exposure to environmental polysaccharides, AHA are not detectable in newborns. At 3–6 months, AHA can first be demonstrated. Titres reach 90% of adult values at 3 years of age and increase to a maximum between 5 and 10 years of age 24–26. Although this test is mentioned in many immunology reviews and textbooks 16,27–30, there are no data on the diagnostic value of AHA for polysaccharide antibody deficiency (PsAD). Only anecdotal reports of low or absent AHA in PID exist (common variable immunodeficiency 27,31, ataxia teleangiectasia 32, Wiskott–Aldrich syndrome 23,33, selective IgM deficiency 34,35, intractable diarrhea of infancy 36, NEMO deficiency 10,37, interleukin-1 receptor-associated kinase 4 (IRAK-4) deficiency 38 and Shwachman–Diamond syndrome 39). Furthermore, there is no consensus on which immunoglobulin is relevant (IgM, IgG or both) or on the assessment technique of AHA.

As AHA titration is clearly more practical and less invasive than pneumococcal antibody response testing, the aim of this study was to determine the diagnostic value of AHA testing in comparison to the pneumococcal antibody response in a retrospective cohort of patients with suspected primary immunodeficiency. To this end, receiver operating characteristic (ROC) curves, sensitivity and specificity of AHA were calculated using pneumococcal antibody response as the gold standard. Furthermore, the clinical significance of low AHA was evaluated by comparing the clinical characteristics of subjects with low Pn antibody response to subjects with low AHA or both normal and by multiple logistic regression.

Methods

Study design

Using the laboratory information software, all patients who underwent both pneumococcal antibody testing and AHA analysis in the University Hospitals Leuven clinical laboratory from January 2008 to November 2012 were identified retrospectively (n = 252). Subjects were excluded when they had been vaccinated with a conjugated vaccine instead of UPV before pneumococcal antibody test (n = 20), when antibody titres prior to vaccination had not been assessed (n = 44), when they had blood group AB (n = 5) or when the subject's blood group was unknown (n = 2). This study was approved by the ethical committee of University Hospitals Leuven.

Laboratory analyses

Anti-pneumococcal polysaccharide IgG titres were assessed by a third-generation World Health Organization (WHO) ELISA, incorporating absorption of samples with cell wall polysaccharides (CPS) and capsular polysaccharide 22F 13. Classically, antibody titres were assessed for serotypes 3, 4 and 9N. In seven subjects, anti-pneumococcal IgG for serotype 18C and 19F had also been determined. An adequate response to UPV was defined as a post-vaccination serotype-specific antibody concentration above 1·3 μg/ml and a twofold or higher increase in antibody titre over the pre-vaccination concentration for more than 70% of serotypes tested (50% for children 2–5 years) 12. When the pre-vaccination titre was above 4 μg/ml, the response was judged adequate even when there was no twofold increase. If a patient aged 6 years or older was responsive to two of serotypes (67%), three additional serotypes were tested for appropriate classification (PS 18C, 19F and 8) (see Supporting information). For clarity, we will use the abbreviation PsAD for a polysaccharide antibody deficiency based on this test, independent of the underlying aetiology or condition. AHA IgG and IgM titres were determined at the immunohaematology laboratory of the Belgian Red Cross at the University Hospitals Leuven, using column agglutination technology (Bio-Rad®) with LISS Coombs columns and neutral columns, respectively.

Medical record review

The medical records of all included subjects were reviewed for upper respiratory tract infections (URTI), lower respiratory tract infections (LRTI), bronchiectasis and invasive infections. More than one LRTI in the history was considered relevant for this study. Bronchiectasis was diagnosed by high-resolution chest CT only (bronchial/arterial diameter > 1). Additionally, total levels of IgG and IgM at the time of pneumococcal antibody production testing were collected for every patient. IgG or IgM hypogammaglobulinaemia was defined as a titre below the age-specific reference value from the laboratory on at least two occasions. IgG or IgM hypergammaglobulinaemia was defined as a titre above the age-specific reference value on two or more occasions. Protein antibody response was registered when available. Protein antibody deficiency was defined as an insufficient response to more than one protein vaccine antigen.

Data analysis

ROC curves were calculated by spss version 20. The clinical relevance of low AHA was investigated by two methods. First, infectious complications were compared in subjects with only low Pn antibody response (group A), only low AHA (group B), both abnormal (group C) or both normal (group D). For the allocation to subgroups, AHA were interpreted as abnormal when any one of the AHA that should be present according to the blood group was below 1/8. The relatively low cut-off of 1/8 was chosen to analyse and compare the clinical characteristics of a population with frank AHA deficiency. Kruskal–Wallis test, Dunn's post-test, χ2 and Fisher's exact tests (for n < 5) for comparison were calculated using the GraphPad Prism software version 5·0 for Windows. Secondly, the association between clinical characteristics and an abnormal test result was studied by logistic regression on the whole cohort (calculated by spss version 20). Because there is no consensus on how to interpret AHA, two strategies were used to study the association between clinical characteristics and low AHA by logistic regression: (1) abnormal AHA = one value below 1/8 (e.g. IgM but not IgG); and (2) abnormal AHA = all values below 1/8 (IgG and IgM for anti-A and anti-B). The threshold of significance was set at a P-value below 0·05.

Results

Study population

A total of 180 subjects were included (105 male, 75 female). All subjects were from Caucasian descent. The median age at Pn antibody testing was 4 years (range 18 months to 51 years, quartiles 1 and 3: 3 and 6 years). The age distribution for AHA testing was comparable to the Pn antibody test (median 4 years, range 12 months to 51 years, Q1 and Q3: 3 and 6 years).

Of the included subjects, 86 had blood group A (47·8%), 14 had blood group B (7·8%) and 80 had blood group O (44·4%).

Nineteen subjects (10·6%) had PsAD based on their response to UPV. The test results and characteristics of these patients are summarized in the Supporting information, Table S1.

Diagnostic accuracy of AHA

To evaluate the accuracy of low AHA as a diagnostic method to detect PsAD, ROC curves were calculated. In white people, anti-B are known to be lower on average than anti-A 24,40. Therefore, separate ROC curves were generated for anti-A and anti-B AHA. The accuracy of IgG as well as IgM AHA was investigated. The ROC curves are shown in Fig. 1. Table 1 summarizes the test populations and the test results. For all AHA, the area under the curve was approximately 0·5 (P = 0·34–1·00). These results demonstrate that none of the AHA are capable of distinguishing subjects with a PsAD accurately.

Fig 1.

Fig 1

Open in a new tab

Receiver operating characteristic curves for allohaemagglutinins to detect a polysaccharide antibody deficiency determined by abnormal pneumococcal antibody response.

Table 1.

Area under the curve with standard error (s.e.), 95% confidence interval (CI) and P-value for receiver operating characteristic curves evaluating allohemagglutinins to detect a polysaccharide antibody deficiency (PsAD)

Measured antibody Number of A or B subjects Number of O subjects PsAD/total subjects AUC s.e. 95% CI P-value
Anti-A IgG 14 (B) 80 12/94 0·536 0·087 0·364–0·707 0·69
Anti-A IgM 14 (B) 80 12/94 0·554 0·089 0·380–0·729 0·54
Anti-B IgG 86 (A) 80 16/166 0·500 0·082 0·340–0·660 1·00
Anti-B IgM 86 (A) 80 16/166 0·573 0·083 0·410–0·736 0·34
Open in a new tab

AUC = area under the curve; Ig = immunoglobulin.

Next, we calculated the sensitivity and specificity of AHA for multiple cut-off values, as no generally accepted cut-offs are available. Most laboratories use a cut-off of 1/32 or 1/16. PID specialists tend to use 1/8 for both below the age of 3 years and 1/16 above 3 years 28,34,38. Sensitivity and specificity of AHA to diagnose PsAD, as defined by the UPV response, are shown in Table 2. The sensitivity was low for all the above-mentioned cut-offs. The likelihood ratio varied between 0·75 and 2·28. The PPV was 25% or lower for all the presented cut-offs. NPVs were approximately 90% independent of the chosen cut-off. Subgroup analyses according to age (< 3 years, 3–17 years and ≥ 18 years) showed the same results in the group of 3–17-year-olds. The groups < 3 years (n = 39) and ≥ 18 years (n = 13) were too small to draw conclusions (see Supporting information, Tables S2–S4)

Table 2.

Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and likelihood ratio (LR) of allohaemagglutinins for detection of a polysaccharide antibody deficiency

Anti-A IgG (n = 94)† Anti-A IgM (n = 94)†
Cut-off Sensitivity Specificity PPV NPV LR Cut-off Sensitivity Specificity PPV NPV LR
< 1/4 16·67 87·80 16·67 87·80 1·37 < 1/4 8·33 96·34 25·00 87·78 2·28
< 1/8 33·33 74·39 16·00 88·41 1·30 < 1/8 16·67 90·24 20·00 88·10 1·71
< 1/16 41·67 58·54 12·82 87·27 1·01 < 1/16 33·33 76·83 17·39 88·73 1·44
< 1/32 58·33 41·46 12·73 89·18 1·00 < 1/32 33·33 63·41 11·76 88·73 0·91
Anti-B IgG (n = 166)‡ Anti-B IgM (n = 166)‡
Cut-off Sensitivity Specificity PPV NPV LR Cut-off Sensitivity Specificity PPV NPV LR
< 1/4 37·50 68·67 11·32 91·15 1·20 < 1/4 37·50 80·67 17·14 92·37 1·94
< 1/8 43·75 55·33 9·46 90·22 0·98 < 1/8 37·50 72·00 12·50 91·53 1·34
< 1/16 43·75 42·00 7·45 87·50 0·75 < 1/16 37·50 58·00 8·70 89·69 0·89
< 1/32 68·75 29·33 9·40 89·80 0·97 < 1/32 62·50 45·33 10·87 91·89 1·14
Open in a new tab
†

Included subjects are blood type B (n = 14) or blood type O (n = 80).

‡

Included subjects are blood type A (n = 86) or blood type O (n = 80). Ig = immunoglobulin.

Clinical characteristics

The number of subjects and characteristics for each subgroup according to AHA and Pn antibody response are shown in Table 3. The age distribution, total IgG and total IgM varied significantly between the four groups. Both PsAD groups (A and C) had a higher median age, but their total IgM and IgG did not differ significantly from the other groups. Dunn's multiple comparison test showed a significantly lower total IgM and total IgG in the group with low AHA versus the group with normal results on both tests. Also, hypoIgM was more frequent in the group with low AHA when compared to the group with both normal (P = 0·004). A higher prevalence of hypoIgG was found in the group with low Pn antibody response (A) and the group with low AHA (B) when compared to group D (normal Pn antibody response and normal AHA) (Fisher's exact test: P = 0·032 and 0·033, respectively). The prevalence of hypergammaglobulinaemia (IgG or IgM) and protein antibody deficiency did not differ significantly between groups.

Table 3.

Characteristics of four subgroups defined by pneumococcal polysaccharide antibody response (Pn antibody response) and allohaemagglutinins (AHA)

Low Pn antibody response only (group A) Low AHA only (group B) Low Pn antibody response and low AHA (group C) Normal Pn antibody response and normal AHA (group D) P-value†
Total no. of subjects 11 77 8 84
Male 8 (72·7%) 50 (64·9%) 3 (37·5%) 44 (52·4%)
Female 3 (27·3%) 27 (35·1%) 5 (62·5%) 40 (47·6%)
Median age in years (Q1–Q3) 7 (5–16·5) 3 (2–5) 8 (4·5–23) 4 (3–6) < 0·001
Median total IgG (mg/l) (Q1–Q3) 8·52 (5·94–9·36) 6·35 (4·9–8·73) 5·49 (4·81–11·85) 8·15 (5·98–9·98) 0·016
Median total IgM (mg/l) (Q1–Q3) 0·75 (0·63–0·99) 0·63 (0·45–0·9) 0·66 (0·32–1·05) 0·93 (0·70–1·18) < 0·001
HypoIgG (no.) 3/11 12/77 2/8 4/84 0·03
HyperIgG (no.) 1/11 7/77 1/8 4/84 0·67
HypoIgM (no.) 0/11 8/77 0/8 084 0·01
HyperIgM (no.) 1/11 0/77 0/8 4/84 0·15
Abnormal protein response (no.) 2/8 4/50 2/6 3/44 0·10
Open in a new tab
†

P-values were calculated by a Kruskal–Wallis test for age, total immunoglobulin (Ig)G and total IgM and by χ2 test for the dichotomous variables (hypoIgG, hyperIgG, hypoIgM, hyperIgM, abnormal protein response).

*

Significant difference between two subgroups by Dunn's multiple test comparison (for age, IgG, IgM) or by Fisher's exact test (for hypoIgG and hypoIgM). Q1 and Q3 are the first and third quartiles (Tukey's Hinges method).

Table 4 shows the prevalence of infectious complications for each group. No significant difference was demonstrated between the prevalence of URTI or > 1 LRTI in the groups (χ2 test). Bronchiectasis, however, was significantly more common in subjects with low Pn antibody response (5/11; 45·5%) when compared to subjects with only low AHA (1/77; 1·3%) or subjects with normal AHA and normal Pn antibody response (2/84; 2·4%) (P < 0·001). The incidence of invasive infections ranged from 0% (0/8; group C) to 18·2% (14/77; group B), but these differences were not significant.

Table 4.

Prevalence of clinical manifestations per subgroup

Low Pn antibody response Group A (n = 11) Low AHA Group B (n = 77) Low Pn antibody response and low AHA Group C (n = 8) Normal Pn antibody response and normal AHA Group D (n = 84)
Prevalence (n° of cases) Proportion (%) 95% CI proportion Prevalence (n° of cases) Proportion (%) 95% CI Prevalence (no. of cases) Proportion (%) 95% CI proportion Prevalence (no. of cases) Proportion (%) 95% CI proportion P-value*
URTI 4 36·4 15·0–64·8 54 70·1 59·1–79·2 4 50 21·5–78·5 58 69·1 58·5–78 0·10
LRTI > 1 6 54·6 28·0–78·8 32 41·6 31·2–52·7 4 50 21·5–78·5 31 36·9 27·4–47·6 0·64
Bronchiectasis 5 45·5 21·3–72·0 1 1·3 < 0·0001–7·7 2 25 6·3–59·9 2 2·4 0·2–8·8 < 0·001
Invasive infections 1 9·1 < 0·0001–39·9 14 18·2 11·0–28·4 0 0 0–37·2 12 14·3 8·2–23·5 0·50
Open in a new tab
*

P-values for difference in proportion between subgroups, calculated by χ2 test. AHA = allohaemagglutinins; CI = confidence interval; Pn antibody response = pneumococcal antibody response; URTI = upper respiratory tract infections; LRTI = lower respiratory tract infections.

Multiple logistic regression revealed a 46·2 times increased risk for bronchiectasis in subjects with an abnormal Pn antibody response when corrected for age (P < 0·001) (Table 5). An increased risk for bronchiectasis in subjects with low AHA (corrected for age) could not be demonstrated. An association was found between bronchiectasis and hyperIgG (odds ratio 7·13; P = 0·01), but not with hypoIgG, hyperIgM or hypoIgM. No significant associations were found when the same analyses were performed for invasive infections, URTI and > 1 LRTI (see Supporting information, Tables S5–S7).

Table 5.

Multiple logistic regression analysis on the association of bronchiectasis (n = 10) with different diagnostic test results. The first row shows a simple logistic regression analysis for age at assessment of clinical manifestations. The following rows represent the results of a new multiple logistic regression model for each row, including age and the mentioned variable

Variable n Intercept Coefficient Standard error Wald test P-value Odds ratio 95% CI odds ratio
Age – −2·998 0·018 0·035 0·257 0·61 1·02 0·95–1·09
Low Pn AB response 19 −3·559 3·832 0·824 21·612 < 0·001 46·18 9·18–232·35
Abnormal AHA 1† 85 −2·659 −0·748 0·712 1·104 0·29 0·47 0·12–1·91
Abnormal AHA 2‡ 34 −3·161 0·663 0·719 0·851 0·36 1·91 0·47–7·95
HyperIgG 13 −3·361 1·965 0·769 6·523 0·01 7·13 1·58–32·23
HypoIgG 21 −2·981 −0·253 1·091 0·054 0·82 0·78 0·09–6·59
HyperIgM 5 −2·963 −18·376 17 971 < 0·001 1·00 – –
HypoIgM 8 −2·940 −18·388 14 206 < 0·001 1·00 – –
Open in a new tab
†

Allohaemagglutinins (AHA) interpreted as abnormal when at least one of the AHA is below 1/8.

‡

AHA interpreted as abnormal when all the AHA are below 1/8. CI = confidence interval; Pn AB response = pneumococcal polysaccharide antibody response; n = number of subjects positive for the categorical variable; Ig = immunoglobulin.

Discussion

To our knowledge, this is the first study to investigate the value of AHA for the detection of a PsAD. We found that AHA cannot discriminate subjects with a PsAD from subjects with a normal Pn antibody response, even using various cut-offs from 1/4 to 1/32. This is consistent with the report of Cheng and colleagues, mentioning normal AHA titres in 12 of 20 SAD patients 41. However, the accuracy of AHA may be underestimated due to an imperfect gold standard bias, as many practical aspects as well as interpretation of the Pn antibody response test are still debated, and no sensitivity/specificity are currently available 12,13. Theoretically, AHA could be better than Pn antibody response in detecting a clinically relevant PsAD. To overcome this bias, we also investigated the clinical significance of low AHA versus low Pn antibody response. In the absence of structural abnormalities of the airways or ciliary dysfunction, bronchiectasis can be a sign of humoral immunodeficiency 41–43. In our cohort, low Pn antibody response but not low AHA was associated with significantly higher prevalence of bronchiectasis. Multiple logistic regression analysis showed a very strong association between Pn antibody deficiency and bronchiectasis but not between low AHA and bronchiectasis. Furthermore, an association between hyperIgG and bronchiectasis was found in this population. The hyperIgG may be due to ongoing infection or inflammation or to a qualitative antibody impairment. In accordance with the previous finding that invasive infections were not characteristic for PsAD 3, we found no significant difference in the prevalence of invasive infections between subjects with low Pn antibody response, low AHA or all normal tests; nor did multiple logistic regression show an association between invasive infections and PsAD or low AHA. Thus, low Pn antibody response but not low AHA was associated with clinically significant PsAD. The NPV of normal AHA titres is approximately 90% but the PPV of low AHA is low; therefore, in our opinion, a Pn antibody response test should be performed in all patients with suspected PsAD.

As well as the lack of correlation with PsAD, AHA testing has a number of other disadvantages. The test is useless in subjects with blood group AB, and it has a wide interinstitutional variability when the agglutination method is used. This last issue could be resolved by newer methods using flow cytometry, which provide accurate and reproducible results with minimal interinstitutional variability 44. Nevertheless, very little is known on these anti-blood group antibodies, and any recent research on AHA has been focused upon ABO-incompatible transplantation (reviewed by Subramanian et al.) 45. Virtually no research was found on age- and race-related normal values of AHA in the healthy population. The most recent publication on normal AHA values in children dates from 1974 25. In healthy children with blood group O, a mean AHA titre above 1/8 was reached at 6–9 months for anti-A and at 12–18 months for anti-B (approximately 10 children tested per age group), but the method of detection did not differentiate between IgM and IgG. According to Stiehm's textbook on immunological disorders, immunologically normal individuals above age 6 months should have a titre of anti-A and/or anti-B IgM of at least 1/8 28. Klein stated that, except in AB subjects, absence of anti-A and anti-B is very rare in healthy individuals 24. Only anecdotal data on AHA titres in PIDs were found in the literature 23,27,31–36,39. More research on AHA in healthy and diseased populations is necessary to define the normal values.

PsAD was found in 10·6% of this study population, which is comparable with the prevalence found in studies in populations with recurrent respiratory tract infections 2–9. However, the low number of ‘true’ positive cases could underestimate the value of AHA. Also, because of the wide interinstitutional variability of detection of AHA, this study should be repeated in other centres or in a multi-centre study to confirm the low correlation between AHA titres and Pn antibody response. The major weakness of this study is its retrospective character and therefore possibly selection bias. Although this study was performed in the population in whom this test is currently proposed for the diagnosis of a polysaccharide antibody deficiency, few included subjects had a genetically confirmed PID known to be associated with PsAD.

To conclude, AHA titres have low sensitivity and specificity in the diagnosis of PsAD. Moreover, there was no association of AHA titre and clinical features such as invasive infections or bronchiectasis. In view of these results, and given the lack of knowledge on normal AHA titres, the diagnostic value of AHA in the evaluation of the polysaccharide antibody response is questionable. Larger prospective studies are needed to establish normal values of AHA and to clarify further the value of low AHA in the work-up of patients with suspected PID.

Acknowledgments

E. V. and H. S. are supported by a PhD fellowship grant of the Research Foundation Flanders (FWO). I. M. is supported by a KOF grant of the KU Leuven – University of Leuven and by the Jeffrey Modell Foundation. G. F., I. M. and X. B. are supported by a GOA grant of the KU Leuven. We thank the laboratory staff from the immunology laboratory of the University Hospitals Leuven for their technical support.

Disclosure

The authors declare no conflicts of interest.

Author contributions

H. S. and I. M. designed the study, analysed the data and prepared the manuscript. E. V., F. V. and X. B. helped with the statistical analyses and reviewed the manuscript. B. V., C. P., G. F. and L. M. reviewed the manuscript. I. M., M. P. and K. D. B. took care of the patients, helped with collection of data and reviewed the manuscript. K. D. V., M. P. E. and X. B. performed the laboratory measurements and reviewed the manuscript.

Supporting Information

Additional Supporting information may be found in the online version of this article at the publisher's web-site:

Table S1. Pneumococcal antibody deficient subjects with clinical presentation, pneumococcal antibody response (Pn antibody response), allohaemagglutinin (AHA) titres and immunological phenotype.

Tables S2–4. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and likelihood ratio (LR) of allohaemagglutinins for detection of a polysaccharide antibody deficiency when the cohort was stratified into three groups according to age < 3 years, 3–17 years and ≥ 18 years.

Table S5. Multiple logistic regression analysis on the association of invasive infections (n = 27) with different diagnostic test results. The first row shows a simple logistic regression analysis for age at assessment of clinical manifestations. The following rows represent the results of a new multiple logistic regression model for each row, including age and the mentioned variable.

Table S6. Multiple logistic regression analysis on the association of upper respiratory tract infections (n = 120) with different diagnostic test results. The first row shows a simple logistic regression analysis for age at assessment of clinical manifestations. The following rows represent the results of a new multiple logistic regression model for each row, including age and the mentioned variable.

Table S7. Multiple logistic regression analysis on the association of lower respiratory tract infections > 1 (n = 73) with different diagnostic test results. The first row shows a simple logistic regression analysis for age at assessment of clinical manifestations. The following rows represent the results of a new multiple logistic regression model for each row, including age and the mentioned variable.

cei0180-0271-sd1.docx (50.8KB, docx)

References

  1. Ambrosino DM, Siber GR, Chilmonczyk BA, Jernberg JB, Finberg RW. An immunodeficiency characterized by impaired antibody responses to polysaccharides. N Engl J Med. 1987;316:790–793. doi: 10.1056/NEJM198703263161306. [DOI] [PubMed] [Google Scholar]
  2. Sanders LA, Rijkers GT, Kuis W, et al. Defective antipneumococcal polysaccharide antibody response in children with recurrent respiratory tract infections. J Allergy Clin Immunol. 1993;91:110–119. doi: 10.1016/0091-6749(93)90303-w. [DOI] [PubMed] [Google Scholar]
  3. Boyle RJ, Le C, Balloch A, Tang ML-K. The clinical syndrome of specific antibody deficiency in children. Clin Exp Immunol. 2006;146:486–492. doi: 10.1111/j.1365-2249.2006.03242.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bossuyt X, Moens L, Van Hoeyveld E, et al. Coexistence of (partial) immune defects and risk of recurrent respiratory infections. Clin Chem. 2007;53:124–130. doi: 10.1373/clinchem.2007.075861. [DOI] [PubMed] [Google Scholar]
  5. Wasserman RL, Sorensen RU. Evaluating children with respiratory tract infections: the role of immunization with bacterial polysaccharide vaccine. Pediatr Infect Dis J. 1999;18:157–163. doi: 10.1097/00006454-199902000-00016. [DOI] [PubMed] [Google Scholar]
  6. Javier FC, 3rd, Moore CM, Sorensen RU. Distribution of primary immunodeficiency diseases diagnosed in a pediatric tertiary hospital. Ann Allergy Asthma Immunol. 2000;84:25–30. doi: 10.1016/S1081-1206(10)62736-6. [DOI] [PubMed] [Google Scholar]
  7. Ruuskanen O, Nurkka A, Helminen M, Viljanen MK, Käyhty H, Kainulainen L. Specific antibody deficiency in children with recurrent respiratory infections: a controlled study with follow-up. Clin Exp Immunol. 2013;172:238–244. doi: 10.1111/cei.12053. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Epstein MM, Gruskay F. Selective deficiency in pneumococcal antibody response in children with recurrent infections. Ann Allergy Asthma Immunol. 1995;75:125–131. [PubMed] [Google Scholar]
  9. Tuerlinckx D, Vermeulen F, Pékus V, et al. Optimal assessment of the ability of children with recurrent respiratory tract infections to produce anti-polysaccharide antibodies. Clin Exp Immunol. 2007;149:295–302. doi: 10.1111/j.1365-2249.2007.03409.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Ku C-L, Dupuis-Girod S, Dittrich A-M, et al. NEMO mutations in 2 unrelated boys with severe infections and conical teeth. Pediatrics. 2005;115:e615–619. doi: 10.1542/peds.2004-1754. [DOI] [PubMed] [Google Scholar]
  11. Sorensen RU, Harvey T, Leiva LE. Selective antibody deficiency with normal immunoglobulins. In: Metodiev K, editor. Immunodeficiency. Rijeka, Croatia: In Tech; 2012. pp. 191–206. [Google Scholar]
  12. Orange JS, Ballow M, Stiehm ER, et al. Use and interpretation of diagnostic vaccination in primary immunodeficiency: a working group report of the Basic and Clinical Immunology Interest Section of the American Academy of Allergy, Asthma and Immunology. J Allergy Clin Immunol. 2012;130(Suppl 3):S1–24. doi: 10.1016/j.jaci.2012.07.002. [DOI] [PubMed] [Google Scholar]
  13. Jeurissen A, Moens L, Raes M, et al. Laboratory diagnosis of specific antibody deficiency to pneumococcal capsular polysaccharide antigens. Clin Chem. 2007;53:505–510. doi: 10.1373/clinchem.2006.080051. [DOI] [PubMed] [Google Scholar]
  14. Borgers H, Moens L, Picard C, et al. Laboratory diagnosis of specific antibody deficiency to pneumococcal capsular polysaccharide antigens by multiplexed bead assay. Clin Immunol. 2010;134:198–205. doi: 10.1016/j.clim.2009.10.006. [DOI] [PubMed] [Google Scholar]
  15. Ballow M. Vaccines in the assessment of patients for immune deficiency. J Allergy Clin Immunol. 2012;130:283–4.e5. doi: 10.1016/j.jaci.2012.04.028. [DOI] [PubMed] [Google Scholar]
  16. Bonilla FA, Bernstein IL, Khan DA, et al. Practice parameter for the diagnosis and management of primary immunodeficiency. Ann Allergy Asthma Immunol. 2005;94(5 Suppl 1):S1–63. doi: 10.1016/s1081-1206(10)61142-8. [DOI] [PubMed] [Google Scholar]
  17. Sorensen RU, Leiva LE. Measurement of pneumococcal polysaccharide antibodies. J Clin Immunol. 2014;34:127–128. doi: 10.1007/s10875-013-9977-z. [DOI] [PubMed] [Google Scholar]
  18. Gelfand EW, Ochs HD, Shearer WT. Controversies in IgG replacement therapy in patients with antibody deficiency diseases. J Allergy Clin Immunol. 2013;131:1001–1005. doi: 10.1016/j.jaci.2013.02.028. [DOI] [PubMed] [Google Scholar]
  19. Poolman J, Borrow R. Hyporesponsiveness and its clinical implications after vaccination with polysaccharide or glycoconjugate vaccines. Expert Rev Vaccines. 2011;10:307–322. doi: 10.1586/erv.11.8. [DOI] [PubMed] [Google Scholar]
  20. Paris K, Sorensen RU. Assessment and clinical interpretation of polysaccharide antibody responses. Ann Allergy Asthma Immunol. 2007;99:462–464. doi: 10.1016/S1081-1206(10)60572-8. [DOI] [PubMed] [Google Scholar]
  21. Vale AM, Schroeder HW., Jr Clinical consequences of defects in B-cell development. J Allergy Clin Immunol. 2010;125:778–787. doi: 10.1016/j.jaci.2010.02.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Fried AJ, Bonilla FA. Pathogenesis, diagnosis, and management of primary antibody deficiencies and infections. Clin Microbiol Rev. 2009;22:396–414. doi: 10.1128/CMR.00001-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Parker W, Yu PB, Holzknecht ZE, Lundberg K, Buckley RH, Platt JL. Specificity and function of ‘natural’ antibodies in immunodeficient subjects: clues to B cell lineage and development. J Clin Immunol. 1997;17:311–321. doi: 10.1023/a:1027378716015. [DOI] [PubMed] [Google Scholar]
  24. Klein HG, Anstee DJ. Mollison's blood transfusion in clinical medicine. 11th edn. Malden, MA/Oxford: Blackwell Publishing; 2006. [Google Scholar]
  25. Fong SW, Qaqundah BY, Taylor WF. Developmental patterns of ABO isoagglutinins in normal children correlated with the effects of age, sex, and maternal isoagglutinins. Transfusion. 1974;14:551–559. doi: 10.1111/j.1537-2995.1974.tb04576.x. [DOI] [PubMed] [Google Scholar]
  26. Godzisz J. [Synthesis of natural allohemagglutinins of the ABO system in healthy children aged 3 months to 3 years] Rev Fr Transfus Immuno-Hématologie. 1979;22:399–412. doi: 10.1016/s0338-4535(79)80034-3. [DOI] [PubMed] [Google Scholar]
  27. Rezaei N, Aghamohammadi A, Notarangelo LD. Primary immunodeficiency diseases: definition, diagnosis, and management. 1st edn. Berlin/Heidelberg: Springer; 2008. [Google Scholar]
  28. Stiehm ER, Ochs HD, Winklestein JA, Rich E. Immunologic disorders in infants and children. 5th edn. Philadelphia: Elsevier Saunders; 2004. [Google Scholar]
  29. Ballow M. Primary immunodeficiency disorders: antibody deficiency. J Allergy Clin Immunol. 2002;109:581–591. doi: 10.1067/mai.2002.122466. [DOI] [PubMed] [Google Scholar]
  30. Bellante JA. Pediatric clinics of North America – clinical immunology. Philadelphia: Elsevier Saunders; 1994. [Google Scholar]
  31. Salzer E, Santos-Valente E, Klaver S, et al. B-cell deficiency and severe autoimmunity caused by deficiency of protein kinase C δ. Blood. 2013;121:3112–3116. doi: 10.1182/blood-2012-10-460741. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Krizon SS, Koval'chuk LV. Natural antibodies (isohemagglutinins, heterophilic, antierythrocytic and antibacterial) in children with congenital immunologic deficiency syndromes. Zh Mikrobiol Epidemiol Immunobiol. 1978;4:31–37. [PubMed] [Google Scholar]
  33. Sullivan KE, Mullen CA, Blaese RM, Winkelstein JA. A multiinstitutional survey of the Wiskott–Aldrich syndrome. J Pediatr. 1994;125:876–885. doi: 10.1016/s0022-3476(05)82002-5. [DOI] [PubMed] [Google Scholar]
  34. Belgemen T, Suskan E, Dogu F, Ikinciogullari A. Selective immunoglobulin M deficiency presenting with recurrent impetigo: a case report and review of the literature. Int Arch Allergy Immunol. 2009;149:283–288. doi: 10.1159/000199725. [DOI] [PubMed] [Google Scholar]
  35. Goldstein MF, Goldstein AL, Dunsky EH, Dvorin DJ, Belecanech GA, Shamir K. Pediatric selective IgM immunodeficiency. Clin Dev Immunol. 2008;2008:624850. doi: 10.1155/2008/624850. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Girault D, Goulet O, Le Deist F, et al. Intractable infant diarrhea associated with phenotypic abnormalities and immunodeficiency. J Pediatr. 1994;125:36–42. doi: 10.1016/s0022-3476(94)70118-0. [DOI] [PubMed] [Google Scholar]
  37. Filipe-Santos O, Bustamante J, Haverkamp MH, et al. X-linked susceptibility to mycobacteria is caused by mutations in NEMO impairing CD40-dependent IL-12 production. J Exp Med. 2006;203:1745–1759. doi: 10.1084/jem.20060085. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Picard C, von Bernuth H, Ghandil P, et al. Clinical features and outcome of patients with IRAK-4 and MyD88 deficiency. Medicine. 2010;89:403–425. doi: 10.1097/MD.0b013e3181fd8ec3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Dror Y, Ginzberg H, Dalal I, et al. Immune function in patients with Shwachman–Diamond syndrome. Br J Haematol. 2001;114:712–717. doi: 10.1046/j.1365-2141.2001.02996.x. [DOI] [PubMed] [Google Scholar]
  40. Ichikawa Y. A study of the iso-agglutinin titres in the sera of Australian subjects (white) Jpn J Med Sci Biol. 1959;12:1–7. doi: 10.7883/yoken1952.12.1. [DOI] [PubMed] [Google Scholar]
  41. Cheng YK, Decker PA, O'Byrne MM, Weiler CR. Clinical and laboratory characteristics of 75 patients with specific polysaccharide antibody deficiency syndrome. Ann Allergy Asthma Immunol. 2006;97:306–311. doi: 10.1016/S1081-1206(10)60794-6. [DOI] [PubMed] [Google Scholar]
  42. Tarzi MD, Grigoriadou S, Carr SB, Kuitert LM, Longhurst HJ. Clinical immunology review series: an approach to the management of pulmonary disease in primary antibody deficiency. Clin Exp Immunol. 2009;155:147–155. doi: 10.1111/j.1365-2249.2008.03851.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Hurst JR, Workman S, Garcha DS, Seneviratne SL, Haddock JA, Grimbacher B. Activity, severity and impact of respiratory disease in primary antibody deficiency syndromes. J Clin Immunol. 2014;34:68–75. doi: 10.1007/s10875-013-9942-x. [DOI] [PubMed] [Google Scholar]
  44. Krishnan NS, Fleetwood P, Higgins RM, et al. Application of flow cytometry to monitor antibody levels in ABO incompatible kidney transplantation. Transplantation. 2008;86:474–477. doi: 10.1097/TP.0b013e31817c4c4c. [DOI] [PubMed] [Google Scholar]
  45. Subramanian V, Ramachandran S, Klein C, et al. ABO-incompatible organ transplantation. Int J Immunogenet. 2012;39:282–290. doi: 10.1111/j.1744-313X.2012.01101.x. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S1. Pneumococcal antibody deficient subjects with clinical presentation, pneumococcal antibody response (Pn antibody response), allohaemagglutinin (AHA) titres and immunological phenotype.

Tables S2–4. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and likelihood ratio (LR) of allohaemagglutinins for detection of a polysaccharide antibody deficiency when the cohort was stratified into three groups according to age < 3 years, 3–17 years and ≥ 18 years.

Table S5. Multiple logistic regression analysis on the association of invasive infections (n = 27) with different diagnostic test results. The first row shows a simple logistic regression analysis for age at assessment of clinical manifestations. The following rows represent the results of a new multiple logistic regression model for each row, including age and the mentioned variable.

Table S6. Multiple logistic regression analysis on the association of upper respiratory tract infections (n = 120) with different diagnostic test results. The first row shows a simple logistic regression analysis for age at assessment of clinical manifestations. The following rows represent the results of a new multiple logistic regression model for each row, including age and the mentioned variable.

Table S7. Multiple logistic regression analysis on the association of lower respiratory tract infections > 1 (n = 73) with different diagnostic test results. The first row shows a simple logistic regression analysis for age at assessment of clinical manifestations. The following rows represent the results of a new multiple logistic regression model for each row, including age and the mentioned variable.

cei0180-0271-sd1.docx (50.8KB, docx)

Articles from Clinical and Experimental Immunology are provided here courtesy of British Society for Immunology

RESOURCES